00:08:35 %git 9d3f7082a8e28e3f5d69b42a311a389155c2454a 00:08:35 07Lasty02 * 0.19-a0-1654-g9d3f708: Deincrementalize protection weapons (elliptic) 10(2 years, 10 months ago, 8 files, 11+ 20-) 13https://github.com/crawl/crawl/commit/9d3f7082a8e2 00:08:50 does anyone remember what problems there were with the old "+1 AC on hit"? 00:08:57 was there a concern about dragging around popcorn? 00:38:32 Unstable branch on CRAWL.XTAHUA.COM updated to: 0.24-a0-486-ged8816983f (34) 00:40:02 Stable (0.23) branch on CRAWL.XTAHUA.COM updated to: 0.23.1-82-g5129a83184 00:45:32 New branch created: pull/1101 (1 commit) 13https://github.com/crawl/crawl/pull/1101 00:45:32 03alexjurkiewicz02 {GitHub} 07https://github.com/crawl/crawl/pull/1101 * 0.24-a0-487-g965425d: Unify two loot lines 10(78 seconds ago, 1 file, 1+ 2-) 13https://github.com/crawl/crawl/commit/965425d4eb7c 00:50:48 03alexjurkiewicz02 {amalloy} 07* 0.24-a0-487-g57318f2: Unify two loot lines (#1101) 10(5 minutes ago, 1 file, 1+ 2-) 13https://github.com/crawl/crawl/commit/57318f2ed029 01:02:13 Unstable branch on crawl.kelbi.org updated to: 0.24-a0-487-g57318f2ed0 (34) 03:23:31 Unstable branch on crawl.beRotato.org updated to: 0.24-a0-487-g57318f2 (34) 03:31:02 Fork (stoatsoup) on crawl.kelbi.org updated to: 0.21.2-1007-gdc84e45775 05:46:52 -!- amalloy is now known as amalloy_ 05:54:03 ng-setup.cc:540:9: error: incompatible operand types ('undead_state_type (*)(species_type)' and 'undead_state_type') 05:54:05 ? *species_undead_type : ng.undead_type; 05:54:07 ^ ~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~ 05:54:09 what does this mean? 05:54:36 the full line is `you.undead_modifier = (species_undead != US_ALIVE) ? *species_undead_type : ng.undead_type;` 05:58:07 oh, typo 05:58:13 h8 badly named variables 06:23:20 -!- Mayoi is now known as erkin 08:14:00 Aha, if you want to test the new Desolation fog, don't do it in a game which is already in Desolation so has the old 200x fog generators, me 08:45:11 hm yeah any upgraded game should still have the old behavior, it's set on levelgen 08:51:18 I wish I'd thought of that _before_ going in in my current game 08:53:32 I could add something that would cap the amount of fog machine calls at 10 per machine (as opposed to 200), but it would produce a lot less fog 08:54:46 (right now it does apply that cap but on fog machine creation) 08:56:18 Seems like extra effort, presumably the corpus of in-progress games in Desolations isn't huge 08:57:44 yeah 08:58:01 it does affect the portals too, though that's only 1 machine per level as opposed to 8-12 machines 09:06:58 possibly the thing to do is to give people a few more days to clear existing desolations, since I think that would make it substantially harder, and then add that fix 10:15:23 Do we use C++17? 10:20:30 no 10:20:32 c++11 10:21:20 dang, I really wanted to use structured bindings :p 10:23:41 2017: <|amethyst> some servers need to be updated, but if we do I think we could move to C++14 reasonably soon 10:24:07 probably still true! 10:36:30 Still I think they're new in 17 10:36:43 So a few more years to go 10:50:05 advil: PR #1102 makes the clua travel functions check the known map bounding box so that they act the same as going into map mode 10:50:34 amalloy: ^ is something we were talking about as well if I'm not mistaken 10:51:34 New branch created: pull/1102 (1 commit) 13https://github.com/crawl/crawl/pull/1102 10:51:34 03Gabriel Marks02 07https://github.com/crawl/crawl/pull/1102 * 0.24-a0-488-g80595b8: Check for known map bounds in clua 10(8 minutes ago, 1 file, 41+ 3-) 13https://github.com/crawl/crawl/commit/80595b8a89aa 11:02:24 Also from 2017: I guess c++17 is sometime in the fall maybe? 11:02:24 wishful thinking 11:02:52 the dream isn't dead 11:02:56 just taking a long nap 11:23:07 Stable (0.23) branch on underhound.eu updated to: 0.23.1-82-g5129a83184 11:28:59 What do people think about allowing clua to be executed from map mode using level map keymaps in the same way it's executed with default keymaps? 11:29:41 And possibly passing the player-centered coordinates of the cursor to the clua function to be executed? 13:28:04 Unstable branch on crawl.akrasiac.org updated to: 0.24-a0-487-g57318f2 (34) 13:30:49 -!- amalloy_ is now known as amalloy 13:47:15 gmarks: i don't see anything obviously wrong with being able to execute lua from X. passing it the coordinates seems obviously good, though i'd be inclined to instead pass it a table like {cursor=pos}, so that it's easy to add more things if there's other stuff that we realise those lua functions would want 13:47:36 (plus having to use the name "cursor" makes it more self-documenting) 13:51:44 Ok 13:54:40 What about interlevel cursor position? On the same level it's easy to give the player-relative position of the cursor but on a different level that can't really be done 13:55:33 If I pass a table with either relative pos or short branch description that might solve it 13:55:55 and an interlevel flag or something 13:57:56 as someone who is not an expert in scripting crawl my main question is, how will the called lua take action? does it basically use crawl.sendkeys for everything? 13:58:56 if so maybe you don't need to send a cursor at all; just sending an enter key would take them to the targeted square, for a simple example 14:04:01 Well there are other functions like set_exclude or get_monster_at that need the actual coordinates 14:05:04 It might be better to have a function to get the cursor position if we are in map mode rather than passing the x, y directly to the macro 14:06:11 Something like: 14:07:22 crawl.in_map_mode, crawl.cursor_level, crawl.cursor_pos, crawl.cursor_is_interlevel 14:08:44 ebering: as resident clua expert, thoughts on the above? 14:35:48 sorry I meant you.in_map_mode, you. ... 15:01:30 03Gabriel Marks02 07https://github.com/crawl/crawl/pull/1102 * 0.24-a0-489-gb99d37a: Refactor logic into shared known_map_bounds function 10(2 minutes ago, 4 files, 44+ 58-) 13https://github.com/crawl/crawl/commit/b99d37a6c4f5 15:30:45 #1103 new PR up to run clua from map mode, doesn't do any exposing of map cursor position to the user under the expectation that such exposing will be handled by a future PR 15:33:23 New branch created: pull/1103 (1 commit) 13https://github.com/crawl/crawl/pull/1103 15:33:23 03Gabriel Marks02 07https://github.com/crawl/crawl/pull/1103 * 0.24-a0-488-g0f54d57: Allow running clua from map viewing mode 10(65 minutes ago, 1 file, 12+ 2-) 13https://github.com/crawl/crawl/commit/0f54d571982e 16:00:09 |amethyst: i'm a little puzzled as to the purpose of FixedVector. it seems pretty similar to std::array. i see it was created before std::array existed, so it makes sense that we added, but if we were rewriting crawl from scratch would we just use std::array instead? 16:00:10 i guess the bounds checking in FixedVector is a big plus compared to std::array 16:06:33 <|amethyst> yeah, I think the bounds checking is the only reason one wouldn't replace it 16:08:56 <|amethyst> You'd probably have to change calls to the ... constructor to use initializer lists, but that wouldn't be a big deal 16:11:37 thanks 16:11:51 <|amethyst> oh, and I guess FixedVector::init would need to be a free function, or replaced by direct calls to std::fill 16:12:23 <|amethyst> oh, there's a std::array::fill that does the same thing 16:16:06 <|amethyst> (and if you use at() instead of operator[], you do get bounds checking) 16:17:50 <|amethyst> (though one might prefer getting an assert immediately where you have a backtrace, as opposed to throwing an exception and having the assert happen in our exception handler) 16:42:29 03Gabriel Marks02 07https://github.com/crawl/crawl/pull/1102 * 0.24-a0-489-gc895d63: Refactor logic into shared known_map_bounds function 10(2 hours ago, 4 files, 45+ 58-) 13https://github.com/crawl/crawl/commit/c895d63cd579 16:42:29 03Gabriel Marks02 07https://github.com/crawl/crawl/pull/1103 * 0.24-a0-488-gae4a1f4: Allow running clua from map viewing mode 10(2 hours ago, 1 file, 10+ 2-) 13https://github.com/crawl/crawl/commit/ae4a1f4daf5a 17:12:49 gmarks: I think that's a reasonable approach (the one you sketched, haven't looked at your PR) but it should be in the crawl. namespace and not the you. namespace 17:13:18 you is a lua interfacte to the player object; crawl is an interface to various aspects of crawl state and interface 17:19:54 It was my impression that most of the functions in crawl were geared towards the interface with almost none towards game state 17:20:29 But it's true that the functionality doesn't really fit in you 17:23:30 crawl.game_is_arena!!! 17:23:42 I'm not sure if that one is exposed in c/dlua 17:23:44 it might be dlua only 17:23:50 or not exposed 17:24:20 but in_map_mode is definitely an interface state and not a gameplay state 17:24:30 atrodo (L12 FeCj) ASSERT(m) in 'mon-util.cc' at line 1978 failed. (Zot (ZotDef)) 17:24:34 atrodo (L12 FeCj) ASSERT(m) in 'mon-util.cc' at line 1978 failed. (Zot (ZotDef)) 17:31:27 ebering: fair enough :) 17:42:04 !crashlog atrodo 17:42:10 1. atrodo, XL22 FeCj, T:128168 (milestone): http://dobrazupa.org/morgue/atrodo/crash-atrodo-20140224-023409.txt 17:42:12 ZOTDEF 17:42:45 true 17:54:12 zotdef will always be with us 17:57:06 It would be cool if henzell linked to the line of code with the failing assert 17:57:14 you can do 17:57:22 !source mon-util.cc:1978 17:57:22 https://github.com/crawl/crawl/blob/master/crawl-ref/source/mon-util.cc#L1978 17:57:34 unfortunately, that won't get you the right version 17:57:44 just speaking in general 17:58:31 source command could be modified to support a version but sequell doesn't currently support having multiple checkouts...actually I'm not sure that this is true 17:58:43 couldn't it detect version and link to the correct blob? 17:58:47 I believe it is using a git repo so it's probably not to ohard to add that support 17:59:14 well it could but I don't think it's too important to have Henzell output that link directly either 17:59:44 but I suppose if you want it to have the correct version, henzell should have access to that info 18:00:03 actually I guess that's just capturing stderr from the process or something 18:00:19 hence it doesn't really know anything about the crash 18:00:33 anyhow I'd recommend you use the source command if you want a link 18:00:54 keeping in mind how a crash may not be for trunk (although 95% of the time it is) 18:02:58 -!- aidanh_ is now known as aidanh 18:22:02 Unstable branch on underhound.eu updated to: 0.24-a0-487-g57318f2ed0 (34) 21:31:13 -!- amalloy is now known as amalloy_ 22:12:12 New branch created: pull/1104 (1 commit) 13https://github.com/crawl/crawl/pull/1104 22:12:12 03alexjurkiewicz02 {GitHub} 07https://github.com/crawl/crawl/pull/1104 * 0.24-a0-488-g797ee5a: Default show_more to false in debug builds 10(3 minutes ago, 1 file, 4+ 0-) 13https://github.com/crawl/crawl/commit/797ee5a20948 22:17:00 -!- amalloy_ is now known as amalloy 22:41:35 uh 22:42:03 not sure why 22:42:42 I would then just have to override that 22:43:12 and I'm probably at least 25% of the people who use a fulldebug build ;-) 22:56:03 i wouldn't mind defaulting to a smaller number of rarely-useful prints in fulldebug mode 22:57:11 when i run fulldebug i just want to see a dprf or two, but i have to wade through mountains of builder messages, for example 23:27:45 I find I usually use ctrl-p to read the log rather than needing to press space a lot 23:28:06 especially when testing some extremely spammy parts of the game, like newgame start or level gen 23:32:21 alexjurkiewicz: you know you can press ESC instead of space, to skip many pages of messages? 23:34:19 woah